Abstract: Voice Activity Detection(VAD) algorithm based on multiresolution spectrum is presented. The proposed VAD uses MFCC of multiresolution spectrum and two classical audio parameters as audio feature, and prejudges silence by detection of multi-gate zero cross ratio, and classifies noise and voice by Support Vector Machines(SVM). Experimental results show that the proposed VAD algorithm achieves overall better performance in all SNRs than VAD of G.729b and other VAD based on MFCC, and computation of micro controller unit based on proposed VAD is lower than that based on G.729b VAD.

Key words: Voice Activity Detection(VAD), wavelet, multiresolution spectrum, Support Vector Machines(SVM)

摘要： 针对静音检测提出基于小波变分辨率频谱特征的检测算法。算法采用多门限过零率对静音进行初判，并提取多个语音感觉特征与基于小波变分辨率频谱的Mel频率倒谱系数(MFCC)组合成语音特征，通过二分类支持向量机对该特征进行分类实现静音检测。测试结果表明，该算法在不同信噪比下语音识别正确率高于G.729b, MFCC特征静音检测算法，基于该算法的视频会议服务器运算量低于使用G.729b静音检测算法的视频系统。

关键词: 静音检测, 小波, 变分辨率频谱, 支持向量机

CLC Number: 

• TP391